A Performance Comparison of Machine Learning Algorithms for Arced Labyrinth Spillways

Salazar, Fernando; Crookston, Brian M.

doi:10.3390/w11030544

Cited by 25 publications

(8 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nowadays, to solve nonlinear problems and complex systems, various neural network and Neuro-fuzzy techniques are widely used. Recently, many researchers such as Emiroglu et al [8], Emiroglu and Kisi [9], Ebtehaj et al [10], Khoshbin et al [11], Zaji et al [12], Azimi et al [13,14] and Salazar and Crookston [15] have used different artificial intelligence (AI) algorithms in their works. For instance, Baylar et al [16] examined the aeration of rectangular weirs by means of the adaptive Neuro-fuzzy inference system (ANFIS) network.…”

Section: Introductionmentioning

confidence: 99%

A novel approach using CFD and neuro-fuzzy-firefly algorithm in predicting labyrinth weir discharge coefficient

Shafiei

Najarchi

Shabanlou

2019

J Braz. Soc. Mech. Sci. Eng.

View full text Add to dashboard Cite

In this paper, for the first time, to model the discharge coefficient of labyrinth weirs, the evolutionary firefly algorithm (FFA) is used for optimizing the membership functions of the adaptive neuro-fuzzy inference system (ANFIS). Also, to enhance the performance of the ANFIS and ANFIS-FFA models, the Monte Carlo simulations (MCs) are employed. Additionally, the k-fold cross-validation is utilized for training and testing the methods. Next, some input dimensionless parameters including the Froude number (Fr), the ratio of the head above the weir to the weir height (H T /P), cycle sidewall angle (α), the ratio of length of the weir crest to the channel width (L c /W), the ratio of length of apex geometry to the width of a single cycle (A/w) and the ratio of the width of a single cycle to the weir height (w/P) are determined. After that, seven different models are developed for ANFIS and ANFIS-FFA. Then, by conducting a sensitivity analysis, the superior models (ANFIS-FFA 5 and ANFIS 5) and the most effective input parameter (Froude number) are identified. Moreover, the error distribution results exhibit that about 70% of the superior model results have errors less than 5%. Subsequently, the discharge coefficient is simulated by means of a computational fluid dynamics (CFD) model. Furthermore, a comparison of the CFD model with the ANFIS and ANFIS-FFA models reveals that the ANFIS-FFA model is significantly more accurate. Also, an uncertainty analysis is performed for the CFD, ANFIS and ANFIS-FFA models. Finally, a very simple code calculating the discharge coefficient of labyrinth wires is presented. This code can be easily employed without any knowledge on ANFIS, FFA and prior information about MATLAB.

show abstract

Section: Introductionmentioning

confidence: 99%

A novel approach using CFD and neuro-fuzzy-firefly algorithm in predicting labyrinth weir discharge coefficient

Shafiei

Najarchi

Shabanlou

2019

J Braz. Soc. Mech. Sci. Eng.

View full text Add to dashboard Cite

show abstract

“…This same algorithm was previously used in regression problems in different applications, e.g., to build regression models to predict dam behaviour [24], to interpret the response of dams to seismic loads [25] and to better understand the behaviour of labyrinth spillways [26]. Other fields of application in the water sector include dam safety [27], water quality [28], classification of water bodies [29] or urban flood mapping [30].…”

Section: Random Forests (Rf)mentioning

confidence: 99%

Anomaly Detection in Dam Behaviour with Machine Learning Classification Models

Salazar

Conde

Irazábal

et al. 2021

Water

Self Cite

View full text Add to dashboard Cite

Dam safety assessment is typically made by comparison between the outcome of some predictive model and measured monitoring data. This is done separately for each response variable, and the results are later interpreted before decision making. In this work, three approaches based on machine learning classifiers are evaluated for the joint analysis of a set of monitoring variables: multi-class, two-class and one-class classification. Support vector machines are applied to all prediction tasks, and random forest is also used for multi-class and two-class. The results show high accuracy for multi-class classification, although the approach has limitations for practical use. The performance in two-class classification is strongly dependent on the features of the anomalies to detect and their similarity to those used for model fitting. The one-class classification model based on support vector machines showed high prediction accuracy, while avoiding the need for correctly selecting and modelling the potential anomalies. A criterion for anomaly detection based on model predictions is defined, which results in a decrease in the misclassification rate. The possibilities and limitations of all three approaches for practical use are discussed.

show abstract

“…Therefore, it allows distinguishing between positive and negative correlation between inputs and response. Since previous applications of this index showed some dependency on the random component of the training algorithm (Salazar and Crookston 2019), a ANN model with 300 random samples is trained for 100 times. This reduces the uncertainty and dependency of the Olden index.…”

Section: Pt-1: Pilot Modelmentioning

confidence: 99%

Engaging soft computing in material and modeling uncertainty quantification of dam engineering problems

Hariri-Ardebili

Salazar

2019

Soft Comput

Self Cite

View full text Add to dashboard Cite

Due to complex nature of nearly all infrastructures (and more specifically concrete dams), the uncertainty quantification is an inseparable part of risk assessment. Uncertainties might be propagated in different aspects depending on their relative importance such as epistemic and aleatory, or spatial and temporal. The objective of this paper is to focus on the material and modeling uncertainties, and to couple them with soft computing techniques aiming to reduce the computational burden of the conventional Monte Carlo-based finite element simulations. Several scenarios are considered in which the concrete and foundation material properties, the water level, and the dam geometry are assumed as random variables. Five soft computing techniques (i.e., random forest, boosted regression trees, multi-adaptive regression splines, artificial neural networks, and support vector machines) are employed to predict various quantities of interest based on different training sizes. It is argued that the artificial neural network is the most accurate algorithm in majority of cases, with enough accuracy as to be useful in reliability analysis as a complement to numerical models. The results with 200 samples in the training set are enough for reaching useful accuracy in most cases. For the simple prediction tasks, the results were predicted with less than 1% error. It is observed that increasing the number of input parameters increases the prediction error. The partial dependence plots provided most sensitive variables in dam design, which were consistent with the physics of the problem. Finally, several practical recommendations are provided for future applications.

show abstract

A Performance Comparison of Machine Learning Algorithms for Arced Labyrinth Spillways

Cited by 25 publications

References 38 publications

A novel approach using CFD and neuro-fuzzy-firefly algorithm in predicting labyrinth weir discharge coefficient

A novel approach using CFD and neuro-fuzzy-firefly algorithm in predicting labyrinth weir discharge coefficient

Anomaly Detection in Dam Behaviour with Machine Learning Classification Models

Engaging soft computing in material and modeling uncertainty quantification of dam engineering problems

Contact Info

Product

Resources

About